Sure, manipulating luck in most cases isn't easy, but IMO it's nothing comparing to visually challenged TASing.
I'm talking about these movies where camera fails to follow character or where time-saving glitch screws screen so bad that the only way to learn what's going on is to monitor dozens of memory values and draw the picture in your mind. Maybe it's not really difficult technically, but it sure is frustrating.
Says Swordless 'Link'. Maybe one should step back a little and stop nominating whatever game you're currently trying? They can't all be the hardest.
The TASes I'm 'working' on are actually quite easy. Ha! Buck the trend!!!1
Joined: 9/2/2008
Posts: 289
Location: United Kingdom
Right now:
*Streets of Rage 3 (2 player Hard) The WIPs were awesome but the enemies' health bars and players' weaker attack power (compared to SOR 2) makes the game long and difficult.
*A another 2D Zelda game on the GBA..... No brainer.
Future runs (new TASable emulators):
* 3D GTA runs. Especially GTA IV with the damned Taxis!
* Mario 64 DS: More characters and more stars what's the optimised route (for an 150 star run)?
I can't really think much right now.
Since were getting PS1 titles (which I only got 1 game for it and it's really obvious) and some more N64 titles too (MupenPlus). The GBA has already got most of the good games which has been TASed, other than Minish Cap.....
I would be more interested with the DS games but I need to sort out the OpenGL thingy on this computer unit or else I won't be able to make or watch any WIPs.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
A better question is why is this topic still going? It's pointless, and everybody's perception of what's hard is going to be different and biased.
And how hard something is doesn't mean anything. It's even stated in the rules that a run shalt not be accepted based on how hard it was to make.
This topic is to important as a bicycle-riding fish is to a lesbian.
Joined: 11/18/2006
Posts: 2426
Location: Back where I belong
Comicalflop wrote:
A better question is why is this topic still going? It's pointless, and everybody's perception of what's hard is going to be different and biased.
Quit whining, good discussion never hurt anyone.
The hardest game to TAS (i.e. Come as close to "perfection" as possible) has these, and probably other, characteristics:
*Is in 3D (three planes of motion is harder to control than two)
*Has several different ways to travel, with small to medium differences in acceleration, top velocity, and turning speed.
*Has random lag frames
*Has multiple random number generators controlling everything from enemy placement, to wind simulation, to room transition time.
*Has controls which update input randomly
*Has decent programming, without major (game-breaking) glitches, but has several small oversights and multiple applications of these oversights.
And so on. I don't think this game has been made yet, but if it has, I doubt anyone has been masochistic enough to try and TAS it.
The hardest game to TAS (i.e. Come as close to "perfection" as possible) has these, and probably other, characteristics:
*Is in 3D (three planes of motion is harder to control than two)
*Has several different ways to travel, with small to medium differences in acceleration, top velocity, and turning speed.
*Has random lag frames
*Has multiple random number generators controlling everything from enemy placement, to wind simulation, to room transition time.
*Has controls which update input randomly
*Has decent programming, without major (game-breaking) glitches, but has several small oversights and multiple applications of these oversights.
And so on. I don't think this game has been made yet, but if it has, I doubt anyone has been masochistic enough to try and TAS it.
I'm pretty sure you mean isn't yet TASable. Sonic Adventure 1+2 might fall somewhat under these condictions.
*Is in 3D (three planes of motion is harder to control than two)
*Has several different ways to travel, with small to medium differences in acceleration, top velocity, and turning speed.
*Has random lag frames
*Has multiple random number generators controlling everything from enemy placement, to wind simulation, to room transition time.
*Has controls which update input randomly
*Has decent programming, without major (game-breaking) glitches, but has several small oversights and multiple applications of these oversights.
These also affect the difficulty:
* Choices made earlier can affect the RNG in undesired ways much, much later, and there's no easy way to undo the effect with a short backtracking. If you want to undo the effect, you have to backtrack a lot, and hours of TASing go to waste. This does happen with some games and it can be really irritating.
* There simply are so many possible choices very frequently in the game, that making a fast run becomes more a question of strategy rather than framegrinding, and tools are of little help making the correct choices. I really like the mentioned Arkanoid as an example of this.
Another way to interpret the question - Which game is hardest to TAS (if you want to get published?)
Then, any boring game is near impossible to TAS and make look interesting.... or....
Any game that is highly competed - such as Super Metroid, SMB, Mario64, and others. I can't think of any other game with multiple people working in contest (not concert) to gain the coveted publication. However, a game like those, which have been SO worked over by many people - makes it hard to find a single frame of improvement. Adventures of Lolo 1 might also deserve this title, even though it is a single author battling himself ;)
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Joined: 5/25/2007
Posts: 399
Location: New England
None of the games I've TAS'd qualify for "the most difficult," but here's how I'd rate them, hardest 1st:
1. Metroid -- random lag, significant randomness, horizontal and vertical acceleration, complex & open-ended routing, character moves at different speeds when jumping/running, many intriguing possible exploits with character completely offscreen that require memory watching
2. Lemmings -- mainly because no prior run published made for a lot of time developing techniques, long game, wide possible solution set for most levels, events separated by long periods of time had to be precisely timed
3. Legacy of the Wizard -- everything depends on luck manipulation to get speed scrolls; very complex & open-ended routing;
4. Faxanadu -- complex routing, complex and completely asinine interaction of character running speed with character attacks, optimization of damage-taking, magic usage, and golds collection over long spans required a lot of planning
5. Rygar -- very very buggy; main difficulty was understanding all the different ways to break the game and how to exploit them
This list is not to brag (that's what submission comments are for!), but to try to lay out all the things I've found that made TAS'ing a game difficult. Based on this, my criteria would be:
-the internals of the game are difficult to fathom (i.e. stuff happens for no apparent reason; simple memory address monitoring is not helpful)
-multiple interacting entities must be controlled simultaneously (i.e. 4 player)
-movement is complex (2d vs 3d, acceleration/deceleration/multiple modes of movement with different speeds)
-complex resource gathering/utilization with extensive randomness
-game sequences are open-ended (i.e. levels can be completed in different orders)
-large, complex map
-no prior TAS's of the game
and finally, I'd add:
-everyone hates the game, so noone wants to help or offer encouragement ;)
Here is how i'd make a game hard to TAS.
Use a RNG that actually detects when someone is trying to manipulate it, and counters any long string of lucky events with a corresponding bit of bad luck later that cannot be manipulated away.
Examples:
1) ok, the player has gotten too many first strikes within the past 50 battles, now the enemy will ALWAYS go first until it has evened out.
2) the players criticals have exceeded the number of misses by 20, the player will now miss 20 times in a row.
3) the player has escaped 20 battles sucesfully without a failure. Prepare to die 20 times from unescapable deathtrap battles. (the game will track the losses between games, so once you have died 10 times from this, it will start letting you escape again)
Have actions near the beginning of the game affect luck near the very end. If you get an unfavorable result, you will have to start ALL OVER. And of course you will be unable to cut and paste your old recording because the luck is different due to your earlier actions.
Prevent all sequence breaking. Or worse, make it so that any sequence breaks slow you down overall.
Stealthily insert delays to slow you down if you are going too fast. Have people wondering why the transitions are suddenly getting so slow halfway into the TAS.
I think there is ways to make games TAS proof or unfriendly without jepardising the quality of the game for a regular player or maybe even a speedrunner.
Ways to make it hard and frustrating
1) Excessive lag caused when the screen scrolls quicker than usual
2) Luck Manipulation based on a number of multiple factors, often influnced by meaningless actions from way back.
3) Game crashes or causes a character to die if the boundaries or a wall is breached
4) Contains glitches which are unfavourable
5) Useful glitches which have awkward side effects e.g Character ghosts or screen goes black meaning you cannot properly see the player or enemies, or the player is off screen and has a series of complex actions to do.
6) Potential shortcuts and tricks which are difficult to determine if they're possible or not.
Ways to make the game hard to differenciate from a normal speedrun
1) A frame rule, possibly rounding up to a second
2) Delayed actions e.g. pressing a button repeatedly will cause the same actions to happen regardless of how fast you press it
3) Game is entirely pre-determined, so a speedrunner doesn't have to react to any slight variation in events.
4) Jumping does not cause the player to slow down
5) Levels can be completed entirely by holding one direction i.e. right FTW
6) Forced wait for certain automated events e.g. must wait for an elavator to arrive at the same pre-determined time in order to continue
7) Holding buttons can manually skip through texts and cutscenes
8) Falling gravity is weak, with a slow max speed
Joined: 10/27/2004
Posts: 1977
Location: Making an escape
Mitjitsu wrote:
Ways to make the game hard to differenciate from a normal speedrun
1) A frame rule, possibly rounding up to a second
2) Delayed actions e.g. pressing a button repeatedly will cause the same actions to happen regardless of how fast you press it
3) Game is entirely pre-determined, so a speedrunner doesn't have to react to any slight variation in events.
4) Jumping does not cause the player to slow down
5) Levels can be completed entirely by holding one direction i.e. right FTW
6) Forced wait for certain automated events e.g. must wait for an elavator to arrive at the same pre-determined time in order to continue
7) Holding buttons can manually skip through texts and cutscenes
8) Falling gravity is weak, with a slow max speed
Super Mario Brothers?
A hundred years from now, they will gaze upon my work and marvel at my skills but never know my name. And that will be good enough for me.
for my point about the random number generator, a normal player wouldn't even notice. Only someone who's managed to manipulate away a fair number of encounters, or get themselves an unbelievable string of luck would have this happen to them.
Say you are expected to have about 15 random encounters between the first town and the first dungeon, if you don't manipulate things. You get there with zero. Cool huh? NOw 15 unskippable inescapable random encounters get added to inside the dungeon in addition to the encounters normally produced. Lets say you had only 13? only two get added. and you don't even notice.
ALso a normal player won't be suprized that their good luck doesn't last forever, and expect runs of good and bad luck, and would never even suspect that the game is deliberately inserting runs of bad luck after the runs of good luck.
Sequence breaks can easily be checked for, it's just that game designers rarely bother for it, and just assume that no one can pull it off. It's called failing to write "can't happen" code. Of course sequence breaking is fun, but shouldn't be required to enjoy the game. But there's no reason they can't just make the game say "very clever, now go back and get what you were supposed to". (oddly enough, metroid fusion actually did something like that!)
If someone pulls this off right, then speedrunning will not be interfered with, and TASing will get an improvement of maybe one minute over a normal speedrun, as that game pushes back and actively hinders your TAS attempt when it's doing things inhumanely fast.
And depending on the game design it can be easy to sneak in delay frames at transitions. The end result of this will be to simply declare "you cannot beat the game before X frames." and if you some how get ahead, frames are evenly added to all remaining transitions to push you back above the minimum frame limit.
I don't think game companies are anti-TAS or anti-TASer any more than they are anti-cheater, but who knows?
Doing things like skewing luck is very risky as if it is discovered (maybe by cheaters), it will alienate legit players. A fundamental (unwritten) rule of luck-based games is that luck must be fair, and when it skews luck like forcing random encounters in inverse proportion to the number of random encounters already done, it is unfair.
Determining results a long time ago is a safe way of handling it. This isn't specifically TASers that are targeted. It's just a way of preventing save+reset abuse, which is quite a common action.
Things like lag or frame rule are not likely to be implemented through intent, just ignorance.
Occasionally, things do pop up that do not affect normal players but are detrimental to TASers, such as rigid RNGs (they may not even affect speedrunners since they use save+reset). For example, Pokemon Mystery Dungeon Red.
There are games which are genuinely challenging to TAS (some 3D games for example) and there are games which are hard to TAS for all the wrong reasons, but not because of malicious intent. I don't recall a game that locks you out near the end for going too fast, but I can't say there will never be such a game.
Use a RNG that actually detects when someone is trying to manipulate it, and counters any long string of lucky events with a corresponding bit of bad luck later that cannot be manipulated away.
You are thinking too complicated. There's no need to do anything like that to completely thwart luck manipulation. Preventing luck manipulation is trivial:
Just have a RNG stream for some purpose (eg. deciding critical hits), which you seed at the beginning of the game (if the system doesn't have a clock which can be used to randomly seed it, then seed it from user input, eg. the first 20 key presses and their timing or whatever), and then just use that RNG stream for that purpose from that point forward.
The only thing the TASer can affect is how the RNG stream is initially seeded, but after that he has absolutely no control. The initial seeding basically fixes all future events. For example, if you have one RNG seed for all critical hits, then the initial seeding fixes the series of critical hits which will happen during the entire game, and there's nothing the player can do to affect it. (If the RNG is of enough quality, the player will be none the wiser about how the critical hits are decided.)
AFAIK a few old console games actually do just that. What I'm surprised about is that this wasn't way more popular. I suppose back in the 80's high-quality, fast-and-lightweight RNGs with large periods were not well-known among game developers, so they resorted to the only thing they could think of: Scramble the user input and use that as the source of randomness.
Prevent all sequence breaking. Or worse, make it so that any sequence breaks slow you down overall.
Sequence breaks are usually caused by programming or level design mistakes and thus are unintentional. Humans are fallible, and such bugs are usually inevitable.
Writing code which tries to detect level design errors being abused can be difficult and bug-prone in itself.
Just have a RNG stream for some purpose (eg. deciding critical hits), which you seed at the beginning of the game (if the system doesn't have a clock which can be used to randomly seed it, then seed it from user input, eg. the first 20 key presses and their timing or whatever), and then just use that RNG stream for that purpose from that point forward.
If that system was implemented it would probably mean that the TASer will be save/reseting a lot just to get the favourable streams of luck.
If that system was implemented it would probably mean that the TASer will be save/reseting a lot just to get the favourable streams of luck.
Yes, you are correct.
However, on a greater scale (such as being forced to reset every 20 seconds or less of gameplay to manipulate all actions), you can't TAS what you want without making it look like a mess. The only solution would be to lower your standards.
If that system was implemented it would probably mean that the TASer will be save/reseting a lot just to get the favourable streams of luck.
If you were making an "un-TASable" game, then you would store the current RNG stream states in the save data, so saving, resetting and loading would not be of any help. :P
